Small-Scale Urban Object Anomaly Detection Using Convolutional Neural Networks with Probability Estimation
Anomaly detection in sequences is a complex problem in security and surveillance. With the exponential growth of surveillance cameras in urban roads, automating them to analyze the data and automatically identify anomalous events efficiently is essential. This paper presents a methodology to detect...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/16/7185 |
| _version_ | 1797583314557599744 |
|---|---|
| author | Iván García-Aguilar Rafael Marcos Luque-Baena Enrique Domínguez Ezequiel López-Rubio |
| author_facet | Iván García-Aguilar Rafael Marcos Luque-Baena Enrique Domínguez Ezequiel López-Rubio |
| author_sort | Iván García-Aguilar |
| collection | DOAJ |
| description | Anomaly detection in sequences is a complex problem in security and surveillance. With the exponential growth of surveillance cameras in urban roads, automating them to analyze the data and automatically identify anomalous events efficiently is essential. This paper presents a methodology to detect anomalous events in urban sequences using pre-trained convolutional neural networks (CNN) and super-resolution (SR) models. The proposal is composed of two parts. In the offline stage, the pre-trained CNN model evaluated a large dataset of urban sequences to detect and establish the common locations of the elements of interest. Analyzing the offline sequences, a density matrix is calculated to learn the spatial patterns and identify the most frequent locations of these elements. Based on probabilities previously calculated from the offline analysis, the pre-trained CNN, now in an online stage, assesses the probability of anomalies appearing in the real-time sequence using the density matrix. Experimental results demonstrate the effectiveness of the presented approach in detecting several anomalies, such as unusual pedestrian routes. This research contributes to urban surveillance by providing a practical and reliable method to improve public safety in urban environments. The proposed methodology can assist city management authorities in proactively detecting anomalies, thus enabling timely reaction and improving urban safety. |
| first_indexed | 2024-03-10T23:36:11Z |
| format | Article |
| id | doaj.art-f492c4e78b6d410fb28fa5427808dcc7 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T23:36:11Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-f492c4e78b6d410fb28fa5427808dcc72023-11-19T02:58:00ZengMDPI AGSensors1424-82202023-08-012316718510.3390/s23167185Small-Scale Urban Object Anomaly Detection Using Convolutional Neural Networks with Probability EstimationIván García-Aguilar0Rafael Marcos Luque-Baena1Enrique Domínguez2Ezequiel López-Rubio3Department of Computer Languages and Computer Science, University of Málaga, Bulevar Louis Pasteur 35, 29071 Málaga, SpainDepartment of Computer Languages and Computer Science, University of Málaga, Bulevar Louis Pasteur 35, 29071 Málaga, SpainDepartment of Computer Languages and Computer Science, University of Málaga, Bulevar Louis Pasteur 35, 29071 Málaga, SpainDepartment of Computer Languages and Computer Science, University of Málaga, Bulevar Louis Pasteur 35, 29071 Málaga, SpainAnomaly detection in sequences is a complex problem in security and surveillance. With the exponential growth of surveillance cameras in urban roads, automating them to analyze the data and automatically identify anomalous events efficiently is essential. This paper presents a methodology to detect anomalous events in urban sequences using pre-trained convolutional neural networks (CNN) and super-resolution (SR) models. The proposal is composed of two parts. In the offline stage, the pre-trained CNN model evaluated a large dataset of urban sequences to detect and establish the common locations of the elements of interest. Analyzing the offline sequences, a density matrix is calculated to learn the spatial patterns and identify the most frequent locations of these elements. Based on probabilities previously calculated from the offline analysis, the pre-trained CNN, now in an online stage, assesses the probability of anomalies appearing in the real-time sequence using the density matrix. Experimental results demonstrate the effectiveness of the presented approach in detecting several anomalies, such as unusual pedestrian routes. This research contributes to urban surveillance by providing a practical and reliable method to improve public safety in urban environments. The proposed methodology can assist city management authorities in proactively detecting anomalies, thus enabling timely reaction and improving urban safety.https://www.mdpi.com/1424-8220/23/16/7185anomaly detectionconvolutional neural networksuper-resolution |
| spellingShingle | Iván García-Aguilar Rafael Marcos Luque-Baena Enrique Domínguez Ezequiel López-Rubio Small-Scale Urban Object Anomaly Detection Using Convolutional Neural Networks with Probability Estimation Sensors anomaly detection convolutional neural network super-resolution |
| title | Small-Scale Urban Object Anomaly Detection Using Convolutional Neural Networks with Probability Estimation |
| title_full | Small-Scale Urban Object Anomaly Detection Using Convolutional Neural Networks with Probability Estimation |
| title_fullStr | Small-Scale Urban Object Anomaly Detection Using Convolutional Neural Networks with Probability Estimation |
| title_full_unstemmed | Small-Scale Urban Object Anomaly Detection Using Convolutional Neural Networks with Probability Estimation |
| title_short | Small-Scale Urban Object Anomaly Detection Using Convolutional Neural Networks with Probability Estimation |
| title_sort | small scale urban object anomaly detection using convolutional neural networks with probability estimation |
| topic | anomaly detection convolutional neural network super-resolution |
| url | https://www.mdpi.com/1424-8220/23/16/7185 |
| work_keys_str_mv | AT ivangarciaaguilar smallscaleurbanobjectanomalydetectionusingconvolutionalneuralnetworkswithprobabilityestimation AT rafaelmarcosluquebaena smallscaleurbanobjectanomalydetectionusingconvolutionalneuralnetworkswithprobabilityestimation AT enriquedominguez smallscaleurbanobjectanomalydetectionusingconvolutionalneuralnetworkswithprobabilityestimation AT ezequiellopezrubio smallscaleurbanobjectanomalydetectionusingconvolutionalneuralnetworkswithprobabilityestimation |